Related Experiment Video
Updated: Aug 5, 2026
Multiparametric Tumor Organoid Drug Screening Using Widefield Live-Cell Imaging for Bulk and Single-Organoid Analysis
Published on: December 23, 2022
Deep transfer learning of cancer drug responses by integrating bulk and single-cell RNA-seq data
Junyi Chen1, Xiaoying Wang2, Anjun Ma3,4
1Department of Biomedical Informatics, College of Medicine, The Ohio State University, Columbus, OH, 43210, USA.
Abstract:
Drug screening data from massive bulk gene expression databases can be analyzed to determine the optimal clinical application of cancer drugs. The growing amount of single-cell RNA sequencing (scRNA-seq) data also provides insights into improving therapeutic effectiveness by helping to study the heterogeneity of drug responses for cancer cell subpopulations. Developing computational approaches to predict and interpret cancer drug response in single-cell data collected from clinical samples can be very useful. We propose scDEAL, a deep transfer learning framework for cancer drug response prediction at the single-cell level by integrating large-scale bulk cell-line data. The highlight in scDEAL involves harmonizing drug-related bulk RNA-seq data with scRNA-seq data and transferring the model trained on bulk RNA-seq data to predict drug responses in scRNA-seq. Another feature of scDEAL is the integrated gradient feature interpretation to infer the signature genes of drug resistance mechanisms. We benchmark scDEAL on six scRNA-seq datasets and demonstrate its model interpretability via three case studies focusing on drug response label prediction, gene signature identification, and pseudotime analysis. We believe that scDEAL could help study cell reprogramming, drug selection, and repurposing for improving therapeutic efficacy.
Insights
scDEAL is a new computational framework that predicts cancer drug responses in single-cell RNA sequencing data. It uses deep transfer learning to analyze bulk cell-line data, improving cancer therapy selection and drug repurposing.
Area of Science:
- Computational biology
- Genomics
- Cancer research
Background:
- Drug screening and gene expression databases offer insights into cancer drug efficacy.
- Single-cell RNA sequencing (scRNA-seq) data reveals heterogeneity in cancer cell drug responses.
- Computational methods are needed to predict and interpret drug responses in single-cell clinical data.
Purpose of the Study:
- To develop a deep transfer learning framework (scDEAL) for predicting cancer drug response at the single-cell level.
- To integrate large-scale bulk cell-line data with scRNA-seq data for enhanced prediction accuracy.
- To provide interpretable insights into drug resistance mechanisms using feature interpretation.
Main Methods:
- scDEAL framework integrates bulk RNA-seq data with scRNA-seq data.
- A deep transfer learning model is trained on bulk data and applied to scRNA-seq data.
- Integrated gradient feature interpretation is used to identify key genes associated with drug resistance.
Main Results:
- scDEAL was benchmarked on six scRNA-seq datasets, demonstrating its predictive capabilities.
- Model interpretability was shown through case studies on drug response prediction, gene signature identification, and pseudotime analysis.
- The framework successfully predicted drug responses and identified resistance mechanisms.
Conclusions:
- scDEAL facilitates accurate prediction of cancer drug response from single-cell data.
- The framework aids in understanding cell reprogramming and identifying potential drug targets.
- scDEAL supports improved drug selection, repurposing, and overall therapeutic efficacy in cancer treatment.
